Abstract
OBJECTIVES: In Guinea, around 17 new cases of HIV occurred each day and it was responsible for 10 deaths a day in 2022. In addition to this burden, regional disparities have emerged over the years. This study aimed to describe and explain the uneven distribution of HIV infection in Guinea using spatial analysis. DESIGN: This is a retrospective cross-sectional secondary analysis using data from the 2012 and 2018 Guinea Demographic and Health Survey (DHS). SETTING: This study was conducted in Guinea. PARTICIPANTS AND METHODS: We conducted a secondary analysis of data from 300 and 400 enumeration areas, respectively, included in the 2012 and 2018 DHS Program for participants aged 15 to 49 who underwent HIV testing. Spatial analysis methods, including Moran I, interpolation and Kulldorff's scan statistic, were applied to examine variation and identify high-risk spatial clusters of HIV prevalence rate. The potential relationship between HIV status and socio-demographic, biological, behavioural and socio-environmental explanatory variables was explored using logistic regression at individual level. RESULTS: In total, 7922 individuals in 2012 and 8539 in 2018 participated in the study. HIV prevalence rate in 2012 and 2018 was 1.9% and 1.5%, respectively. Across Guinea's 33 prefectures, HIV prevalence rate varied from 0% to 3.9% in 2012 and from 0% to 3.5% in 2018. Spatial analysis identified four significant high-risk spatial clusters in 2012 and one high-risk cluster in 2018. The high-risk clusters in 2012 were in Kissidougou (relative risk (RR)=3.97; p value=0.037), Matam (RR=2.80; p value=0.019), Pita (RR=3.46; p value=0.035) and N'zerekore prefectures (RR=6.08; p value=0.027), the high-risk cluster in 2018 was located in Boffa prefecture (RR=3.95; p value=0.022). Factors significantly and positively associated with HIV infection in 2012 included age class 25-34 (aOR: 2.20; 95% CI 1.40 to 3.47), age class 35-49 (aOR: 2.43; 95% CI 1.51 to 3.92), number of HIV healthcare facilities>30 (aOR: 2.14; 95% CI 1.34 to 3.43). HIV infection was significantly lower in men (aOR: 0.52; 95% CI 0.35 to 0.77). In 2018, in addition to age groups 25-34 years (aOR=1.90; 95% CI 1.18 to 3.04) and 35-49 years (aOR=2.25; 95% CI 1.40 to 3.64), the Soussou ethnicity group (aOR=1.73; 95% CI 1.04 to 2.87) was also positively associated with HIV infection. CONCLUSION: This study describes the spatial distribution of HIV prevalence rate and identified high-risk clusters in Guinea. In addition, risk factors associated with HIV status were identified. The information can help prioritise surveillance and response efforts to control HIV in Guinea.